Relocations of Earthquakes (1899–1917) in South-Central Alaska

I have relocated 18 earthquakes occurring in the south-central Alaska region between 1899 and 1917 using a bootstrap relocation technique. Locations of events within the Yakutat region suggest that the 1899 sequence began on 4 September with a M_S = 7.9 event within the area of the Pamplona fault zone/western Transition fault zone, rupturing the western portion of the North American/Pacific plate interface. A M_S = 7.4 event on 10 September appears to have ruptured the offshore portion of the plate interface to the east of the 4 September event. This was followed by a M_S = 8.0 event that likely ruptured the onshore and down-dip portion of the plate interface. A M_S = 7.0 event in 1908 may have ruptured a small portion of the plate interface between the 4 September and 10 September events. Events occurring between 1911 and 1916 in the Prince William Sound region appear to be slab events occurring in similar locations to more recent seismicity. Within the Kodiak region the 1900 earthquake of M_S = 7.7 has a location consistent with the rupture of the Kodiak asperity which also ruptured during the 1964 great Alaska earthquake. Other large magnitude Kodiak events appear to be associated with regions of recent seismicity, including the Karluk Lake area of southwestern Kodiak Island and the Albatross Basin located offshore southeast of Kodiak Island. Space-time seismicity patterns since 1899 indicate that magnitude 6 to7 events have occurred with regularity in the Kodiak Island region; that there has been a lack of magnitude ≥ 6 events in the Prince William Sound region since 1964, and that the Yakutat region has remained notably quiescent at the magnitude ≥ 6 level.     

The rupture process and tectonic implications of the great 1964 Prince William Sound earthquake

We have determined the rupture history of the March 28, 1964, Prince Williams Sound earthquake (M _w=9.2) from long-period WWSSNP-wave seismograms. Source time functions determined from the long-periodP waves indicate two major pulses of moment release. The first and largest moment pulse has a duration of approximately 100 seconds with a relatively smooth onset which reaches a peak moment release rate at about 75 seconds into the rupture. The second smaller pulse of moment release starts at approximately 160 seconds after the origin time and has a duration of roughly 40 seconds. Because of the large size of this event and thus a deficiency of on-scale, digitizableP-wave seismograms, it is impossible to uniquely invert for the location of moment release. However, if we assume a rupture direction based on the aftershock distribution and the results of surface wave directivity studies we are able to locate the spatial distribution of moment along the length of the fault. The first moment pulse most likely initiated near the epicenter at the northeastern down-dip edge of the aftershock area and then spread over the fault surface in a semi-circular fashion until the full width of the fault was activated. The rupture then extended toward the southwest approximately 300 km (Ruff andKanamori, 1983). The second moment pulse was located in the vicinity of Kodiak Island, starting at 500 km southwest of the epicenter and extending to about 600 km. Although the aftershock area extends southwest past the second moment pulse by at least 100 km, the moment release remained low. We interpret the 1964 Prince William Sound earthquake as a multiple asperity rupture with a very large dominant asperity in the epicentral region and a second major, but smaller, asperity in the Kodiak Island region.The zone that ruptured in the 1964 earthquake is segmented into two regions corresponding to the two regions of concentrated moment release. Historical earthquake data suggest that these segments behaved independently during previous events. The Kodiak Island region appears to rupture more frequently with previous events occurring in 1900, 1854, 1844, and 1792. In contrast, the Prince William Sound region has much longer recurrence intervals on the order of 400–1000 years.     

Where and why do large shallow intraslab earthquakes occur?

We try to find how often, and in what regions large earthquakes (M≥7.0) occur within the shallow portion (20-60 km depth) of a subducting slab. Searching for events in published individual studies and the Harvard University centroid moment tensor catalogue, we find twenty such events in E. Hokkaido, Kyushu-SW, Japan, S. Mariana, Manila, Sumatra, Vanuatu, N. Chile, C. Peru, El Salvador, Mexico, N. Cascadia and Alaska. Slab stresses revealed from the mechanism solutions of these large intraslab events and nearby smaller events are almost always down-dip tensional. Except for E. Hokkaido, Manila, and Sumatra, the upper plate shows horizontal stress gradient in the arc-perpendicular direction. We infer that shear tractions are operating at the base of the upper plate in this direction to produce the observed gradient and compression in the outer fore-arc, balancing the down-dip tensional stress of the slab. This tectonic situation in the subduction zone might be realized as part of the convection system with some conditions, as shown by previous numerical simulations.     

The rupture process and asperity distribution of three great earthquakes from long-period diffracted P-waves

Maximum earthquake size varies considerably amongst the subduction zones. This has been interpreted as a variation in the seismic coupling, which is presumably related to the mechanical conditions of the fault zone. The rupture process of a great earthquake indicates the distribution of strong (asperities) and weak regions of the fault. The rupture process of three great earthquakes (1963 Kurile Islands, M_W = 8.5; 1965 Rat Islands, M_W = 8.7; 1964 Alaska, M_W = 9.2) are studied by using WWSSN stations in the core shadow zone. Diffraction around the core attenuates the P-wave amplitudes such that on-scale long-period P-waves are recorded. There are striking differences between the seismograms of the great earthquakes; the Alaskan earthquake has the largest amplitude and a very long-period nature, while the Kurile Islands earthquake appears to be a sequence of magnitude 7.5 events.The source time functions are deconvolved from the observed records. The Kurile Islands rupture process is characterized by the breaking of asperities with a length scale of 40–60 km, and for the Alaskan earthquake the dominant length scale in the epicentral region is 140–200 km. The variation of length scale and M_W suggests that larger asperities cause larger earthquakes. The source time function of the 1979 Colombia earthquake (M_W = 8.3) is also deconvolved. This earthquake is characterized by a single asperity of length scale 100–120 km, which is consistent with the above pattern, as the Colombia subduction zone was previously ruptured by a great (M_W = 8.8) earthquake in 1906.The main result is that maximum earthquake size is related to the asperity distribution on the fault. The subduction zones with the largest earthquakes have very large asperities (e.g. the Alaskan earthquake), while the zones with the smaller great earthquakes (e.g. Kurile Islands) have smaller scattered asperities.     

Sources of Tsunami and Tsunamigenic Earthquakes in Subduction Zones

—We classified tsunamigenic earthquakes in subduction zones into three types earth quakes at the plate interface (typical interplate events), earthquakes at the outer rise, within the subducting slab or overlying crust (intraplate events), and "tsunami earthquakes" that generate considerably larger tsunamis than expected from seismic waves. The depth range of a typical interplate earthquake source is 10–40km, controlled by temperature and other geological parameters. The slip distribution varies both with depth and along-strike. Recent examples show very different temporal change of slip distribution in the Aleutians and the Japan trench. The tsunamigenic coseismic slip of the 1957 Aleutian earthquake was concentrated on an asperity located in the western half of an aftershock zone 1200km long. This asperity ruptured again in the 1986 Andreanof Islands and 1996 Delarof Islands earthquakes. By contrast, the source of the 1994 Sanriku-oki earthquake corresponds to the low slip region of the previous interplate event, the 1968 Tokachi-oki earthquake. Tsunamis from intraplate earthquakes within the subducting slab can be at least as large as those from interplate earthquakes; tsunami hazard assessments must include such events. Similarity in macroseismic data from two southern Kuril earthquakes illustrates difficulty in distinguishing interplate and slab events on the basis of historical data such as felt reports and tsunami heights. Most moment release of tsunami earthquakes occurs in a narrow region near the trench, and the concentrated slip is responsible for the large tsunami. Numerical modeling of the 1996 Peru earthquake confirms this model, which has been proposed for other tsunami earthquakes, including 1896 Sanriku, 1946 Aleutian and 1992 Nicaragua.     

Large earthquake doublets and fault plane heterogeneity in the northern Solomon Islands subduction zone

In the Solomon Islands and New Britain subduction zones, the largest earthquakes commonly occur as pairs with small separation in time, space and magnitude. This doublet behavior has been attributed to a pattern of fault plane heterogeneity consisting of closely spaced asperities such that the failure of one asperity triggers slip in adjacent asperities. We analyzed body waves of the January 31, 1974,M _w =7.3, February 1, 1974,M _w =7.4, July 20, 1975 (1437)M _w =7.6 and July 20, 1975 (1945),M _w =7.3 doublet events using an iterative, multiple station inversion technique to determine the spatio-temporal distribution of seismic moment release associated with these events. Although the 1974 doublet has smaller body wave moments than the 1975 events, their source histories are more complicated, lasting over 40 seconds and consisting of several subevents located near the epicentral regions. The second 1975 event is well modeled by a simple point source initiating at a depth of 15 km and rupturing an approximate 20 km region about the epicenter. The source history of the first 1975 event reveals a westerly propagating rupture, extending about 50 km from its hypocenter at a depth of 25 km. The asperities of the 1975 events are of comparable size and do not overlap one another, consistent with the asperity triggering hypothesis. The relatively large source areas and small seismic moments of the 1974 doublet events indicate failure of weaker portions of the fault plane in their epicentral regions. Variations in the roughness of the bathymetry of the subducting plate, accompanying subduction of the Woodlark Rise, may be responsible for changes in the mechanical properties of the plate interface.To understand how variations in fault plane coupling and strength affect the interplate seismicity pattern, we relocated 85 underthrusting earthquakes in the northern Solomon Islands Are since 1964. Relatively few smaller magnitude underthrusting events overlap the Solomon Islands doublet asperity regions, where fault coupling and strength are inferred to be the greatest. However, these asperity regions have been the sites of several previous earthquakes withM _s 7.0. The source regions of the 1974 doublet events, which we infer to be mechanically weak, contain many smaller magnitude events but have not generated any otherM _s 7.0 earthquakes in the historic past. The central portion of the northern Solomon Islands Arc between the two largest doublet events in 1971 (studied in detail bySchwartz et al., 1989a) and 1975 contains the greatest number of smaller magnitude underthrusting earthquakes. The location of this small region sandwiched between two strongly coupled portions of the plate interface suggest that it may be the site of the next large northern Solomon Islands earthquake. However, this region has experienced no known earthquakes withM _s 7.0 and may represent a relatively aseismic portion of the subduction zone.     

Note on the 1964 Alaska Tsunami Generation by Horizontal Displacements of Ocean Bottom. Numerical Modeling of the Runup in Chenega Cove, Alaska

A numerical model of the wave dynamics in Chenega Cove, Alaska during the historic M _w 9.2 megathrust earthquake is presented. During the earthquake, locally generated waves of unknown origin were identified at the village of Chenega, located in the western part of Prince William Sound. The waves appeared shortly after the shaking began and swept away most of the buildings while the shaking continued. We model the tectonic tsunami in Chenega Cove assuming different tsunami generation processes. Modeled results are compared with eyewitness reports and an observed runup. Results of the numerical experiments let us claim the importance of including both vertical and horizontal displacement into the 1964 tsunami generation process. We also present an explanation for the fact that arrivals of later waves in Chenega were unnoticed.     

Frequency-dependent rupture process,stress change,and seismogenic mechanism of the 25 April 2015 Nepal Gorkha <Emphasis Type="Italic">M</Emphasis><Subscript>w</Subscript> 7.8 earthquake

On 25 April 2015, an M _w 7.8 earthquake occurred on the Main Himalaya Thrust fault with a dip angle of ~ 7° about 77 km northwest of Kathmandu, Nepal. This Nepal Gorkha event is the largest one on the Himalayan thrust belt since 1950. Here we use the compressive sensing method in the frequency domain to track the seismic radiation and rupture process of this event using teleseismic P waves recorded by array stations in North America. We also compute the distribution of static shear stress changes on the fault plane from a coseismic slip model. Our results indicate a dominant east-southeastward unilateral rupture process from the epicenter with an average rupture speed of ~3 km s^?1. Coseismic radiation of this earthquake shows clear frequency-dependent features. The lower frequency (0.05–0.3 Hz) radiation mainly originates from large coseismic slip regions with negative coseismic shear stress changes. In comparison, higher frequency (0.3–0.6 Hz) radiation appears to be from the down-dip part around the margin of large slip areas, which has been loaded and presents positive coseismic shear stress changes. We propose an asperity model to interpret this Nepal earthquake sequence and compare the frequency-dependent coseismic radiation with that in subduction zones. Such frequency-dependent radiation indicates the depth-varying frictional properties on the plate interface of the Nepal section in the main Himalaya thrust system, similar to previous findings in oceanic subduction zones. Our findings provide further evidence of the spatial correlation between changes of static stress status on the fault plane and the observed frequency-dependent coseismic radiation during large earthquakes. Our results show that the frequency-dependent coseismic radiation is not only found for megathrust earthquakes in the oceanic subduction environment, but also holds true for thrust events in the continental collision zone.     

Body-wave wave-form modelling and source parameters for the indochina border earthquakes

Wave-form modelling of body waves has been done to study the seismic source parameters of three earthquakes which occurred on October 21, 1964 (M _b =5.9), September 26, 1966 (M _b =5.8) and March 14, 1967 (M _b =5.8). These events occurred in the Indochina border region where a low-angle thrust fault accommodates motion between the underthrusting Indian plate and overlying Himalaya. The focal depths of all these earthquakes are between 12–37 km. The total range in dip for the three events is 5°–20°. TheT axes are NE-SW directed whereas the strikes of the northward dipping nodal planes are generally parallel to the local structural trend. The total source durations have been found to vary between 5–6 seconds. The average values of seismic moment, fault radius and dislocation are 1.0–11.0×10²⁵ dyne-cm, 7.7–8.4km and 9.4–47.4 cm, respectively whereas stress drop, apparent stress and strain energy are found to be 16–76 bars, 8.2–37.9 bars and 0.1–1.7×10²¹ ergs, respectively. These earthquakes possibly resulted due to the tension caused by the bending of the lithospheric plate into a region of former subduction which is now a zone of thrusting and crustal shortening.     

Relation between the Subducting Plate and Seismicity Associated with the Great 1964 Alaska Earthquake

—Tectonic studies of the great 1964 Alaska earthquake have underappreciated the nature of the subducted plate in influencing seismicity. We compare seismological observations in the Prince William and Kodiak areas that ruptured during this earthquake with the corresponding morphology and structure of the subducting plate. The upper plate geology (Prince William Terrane) and velocity structure are the same in both areas. In the Prince William area where the Yakutat Terrane subducted, the energy released and coupling were stronger than above the Kodiak subduction zone where thick trench sediment subducts. The conjecture that lower plate character or the amount of subducted sediment affects coupling helps explain variability in seismology, geodetic inversions and the horizontal velocity of GPS stations.     

Upper crustal structure of the western Corinth Gulf, greece, inferred from arrival times of the January 2010 earthquake sequence

The western part of the Corinth Gulf attracts attention due to its seismically active fault system and considerable seismic hazard. A moderate size earthquake occurred close to the town of Efpalio on January 18, 2010, followed by a sequence of smaller earthquakes. In the present paper we use this sequence to derive a local structural model for the region in the vicinity of Efpalio. The model is based on the minimization of traveltime residuals. In particular, we used arrival times from 51 selected events recorded on January 19 and 20 by at least 5 stations at epicentral distances less than about 25 km. A variant of the method of conjugate gradients has been used for this purpose. In comparison with several previous models, the new model is characterized by higher velocities to a depth of about 8 km. The velocity ratio in the model is v_P / v_S = 1.83. The hypocentres of the selected earthquakes lay at depths between about 5 and 9 km, but their distribution is rather irregular.     

Numerical Modeling and Observations of Tsunami Waves in Alberni Inlet and Barkley Sound, British Columbia

Alberni Inlet is a long and narrow fjord adjacent to Barkley Sound on the Pacific Coast of Vancouver Island, Canada. Port Alberni, at the head of the inlet, was affected in 1964 by the largest Pacific tsunami waves in Canadian history. We use observations and results from two numerical models to investigate the resonant characteristics of the region and amplification of tsunami waves in Barkley Sound and Alberni Inlet. The first model (A) was forced at its open boundary with a stationary autoregressive signal, similar to the observed background noise. The second model (B) used an initial sea-level deformation from a potential earthquake off California in the southern segment of the Cascadia Subduction Zone, producing transient tsunami waves. Spectral, cross-spectral and frequency-time (f-t) analyses of the observations were used to examine the resonant properties and topographic response of the local area. The respective results show large admittance functions over a wide 0.5–0.9 cph frequency band, implying a low Q factor but high amplification of arriving waves. This unusual behavior is a result of two effects: A quarter-wave resonance of the system for its fundamental Helmholtz mode and amplification due to the narrowing of the channel cross section from Barkley Sound to Alberni Inlet. The model A numerical results agree favorably with the observations, indicating an energetic resonant mode at frequency of ~0.53 cph (112 min), with its nodal line located near the entrance to Barkley Sound and amplification factor value close to 12. The results from the tsunami propagation model (B) yield spectral characteristics similar to those from the model A and from the observations. The maximum tsunami current speed for this scenario is 2.4 ms^?1 in Sproat Narrows, which divides Alberni Inlet into two parts, while the largest computed wave height is 1.6 m in the northern Alberni Inlet, in the area of Port Alberni.     

Coseismic slip in the 1964 Prince William Sound earthquake: A new geodetic inversion

The 1964 Prince William Sound earthquake (March 28, 1964;M _w =9.2) caused crustal deformation over an area of approximately 140,000 km² in south central Alaska. In this study geodetic and geologic measurements of this surface deformation were inverted for the slip distribution on the 1964 rupture surface. Previous seismologic, geologic, and geodetic studies of this region were used to constrain the geometry of the fault surface. In the Kodiak Island region, 28 rectangular planes (50 by 50 km each) oriented 218°N, with a dip varying from 8^o nearest the Aleutian trench to 9^o below Kodiak Island, define the rupture surface. In the Prince William Sound region 39 planes with variable dimensions (40 by 50 km near the trench, 64 by 50 km inland) and orientation (218°N in the west and 270°N in the east) were used to approximate the complex faulting. Prior information was introduced to constrain offshore dip-slip values, the strike-slip component, and slip variation between adjacent planes. Our results suggest a variable dip-slip component with local slip maximums occurring near Montague Island (up to 30 m), further to the east near Kayak Island (up to 14 m), and trenchward of the northeast segment of Kodiak Island (up to 17m). A single fault plane dipping 30°NW, corresponding to the Patton Bay fault, with a slip value of 8 m modeled the localized but large uplift on Montague Island. The moment calculated on the basis of our geodetically derived slip model of 5.0×10²⁹ dyne cm is 30% less than the seismic moment of 7.5×10²⁹ dyne cm calculated from long-period surface waves (Kanamori, 1970) but is close to the seismic moment of 5.9×10²⁹ dyne cm obtained byKikuchi andFukao (1987).     

An unusual zone of seismic coupling in the Bonin arc: The 1972 Hachijo-Oki earthquakes and related seismicity

The 1972 February and December Hachijo-Oki earthquakes (M _s=7.3 and 7.4), in the northernmost part of the Izu-Bonin subduction zone, are the only major events (M _s>7.0) in the Bonin arc for the past 80 years. Relocation of the hypocenters, using one smaller event having a wellconstrained focal depth as a master event, shows that the depth of the February event is 10 km shallower than that of the December event. We have determined the rupture process for both events by minimizing the error in waveform between observed and synthetic seismograms. Although the number of available stations are limited, the depth range of the major energy release for the December event extends deeper than for the February one. The rupture propagated up-dip for both events. It is likely that the rupture zone of the two events overlapped, and that the December event ruptured the deeper part. This suggestion is consistent with the observation that the aftershock zones of both events overlap with that of the December event shifted landward. The waveforms of the December event have a smaller high frequency component than those of the February event, suggesting that the stress at the thrust zone became more uniform or reduced after the February event.No thrust type smaller event occurred near the rupture zone. Instead, theP-axes of smaller events are parallel to the dip of the slab and theirT-axes dip to the southwest. Focal depths of these events estimated byP-wave forward modeling are generally between 40–50 km and located beneath the thrust zone. We thus interpret them as the events within the Pacific slab near the zone ruptured by the two major events. The stress concentration around the rupture zone of the major events is suggested to have triggered these slab events. After the occurrence of the large events, the slab events are concentrated near the deeper portion of the rupture zone. These events may have been caused by the loading of the down-dip compressional stress near the down-dip end of the rupture zone due to the rupture. The occurrence of the doublet of large earthquakes and a number of down-dip compressional events beneath their rupture zones in a shallow portion of the subducting slab indicates an unusual zone of seismic coupling in the Bonin arc, most of which is seismically quiescent.     

Long term monitoring for oil in the Exxon Valdez spill region

In the aftermath of the 1989 Exxon Valdez oil spill, a Long Term Environmental Monitoring Program (LTEMP) has been regularly sampling mussels (and some sediments) for polycyclic aromatic and saturated hydrocarbons (PAH and SHC) at sites in Port Valdez, Prince William Sound, and the nearby Gulf of Alaska region. After 1999, a decreasing trend appears in total PAH (TPAH) in tissues at all sites with current values below 100 ng/g dry weight (many below 50 ng/g). Currently, most samples reflect a predominantly dissolved-phase signal. This new low in TPAH likely represents ambient background levels. Synchrony in TPAH time-series and similarities in the hydrocarbon signatures portray regional-scale dynamics. The five inner Prince William Sound sites show similar composition and fluctuations that are different from the three Gulf of Alaska sites. The two Port Valdez sites represent a unique third region primarily influenced by the treated ballast water discharge from the Alyeska Marine Terminal. Prince William Sound has reverted to a stable environment of extremely low level contamination in which local perturbations are easily detected.     

Using tolerance intervals to assess recovery of mussel beds impacted by the Exxon Valdez oil spill

Polycyclic aromatic hydrocarbons (PAH) have been measured in mussel tissues in early spring and summer since 1993 throughout Prince William Sound (PWS) and the Gulf of Alaska (GOA). Season-specific thresholds were established at reference sites to identify ‘above background’ total PAH levels. Thresholds were estimated using one-sided 99% tolerance limits. Thresholds were similar across reference sites but differed by an order of magnitude across seasons. Trends in total PAH since 1998 were assessed for sites impacted by the 1989 Exxon Valdez oil spill or the Alyeska Marine Terminal. Summer samples exhibited no trends; early spring samples declined. In early spring, all sites were judged ‘recovered’ by 2004; in summer, one site in western Prince William Sound and two in the western GOA exceeded thresholds by 11 ng/g dry weight or less. Robust estimation methods prevented bias from observations affected by unknown releases or laboratory errors.     

Aftershock Activity and Frequency-dependent Low Coda Qc in the Epicentral Region of the 1999 Chamoli Earthquake of Mw 6.4

— On 28 March, 1999 (19:05:10.09, UT) a significant earthquake of M _w 6.4 occurred in the Garhwal Himalaya (30.555°N, 79.424°E). One hundred and ten well-recorded aftershocks show a WNW-ESE trending northeasterly dipping seismic zone extending from a depth of 2 to 20?km. As the main shock hypocenter occurred at the northern end of this seismic zone and aftershocks extended updip, it is inferred that the main-shock rupture nucleated on the detachment plane at a depth of 15?km and then propagated updip along a NE-dipping thrust plane. Further, the epicentral distribution of aftershocks defines a marked concentration near a zone where main central thrust (MCT) takes a significant turn towards the north, which might be acting as an asperity in response to the NNE compression due to the underthrusting of Himalayan orogenic process prevalent in the entire region. Presence of high seismicity including five earthquakes of magnitude exceeding 6 and twelve earthquakes of magnitude exceeding 5 in the 20th century is presumed to have caused a higher level of shallow crustal heterogeneity in the Garhwal Himalaya, a site lying in the central gap zone of the Himalayan frontal arc. Attenuation property of the medium around the epicentral area of the 1999 Chamoli earthquake, covering a circular area of 61,500?km² with a radius of 140?km, is studied by estimating the coda Q _c from 48 local earthquakes of magnitudes varying from 2.5–4.8. These earthquakes were recorded at nine 24-bit REFTEK digital stations; two of which were equipped with three-component CMG40T broadband seismometers and others with three-component L4-3D short-period seismometers. The estimated Q _o values at different stations suggest on average a low value of the order of (30?±?0.8), indicating an attenuating crust beneath the entire region. The frequency-dependent relation indicates a relatively low Q _c at lower frequencies (1–3?Hz) that can be attributed to the loss of energy due to scattering on heterogeneities and/or the presence of faults and cracks. The large Q _c at higher frequencies may be related to the propagation of backscattered body waves through deeper parts of the lithosphere where less heterogeneities are expected. An important observation is that the region north of MCT (more rigid highly metamorphosed crystalline rocks) is less attenuative in comparison to the region south of MCT (less rigid slightly metamorphosed rocks (sedimentary wedge)). The acceleration decays to 50% at 20?km distance and to 7% at 100?km. Hence, even 1g acceleration at the source may not cause significant damage beyond 100?km in this region.     

Rupture process of large earthquakes in the northern Mexico subduction zone

The Cocos plate subducts beneath North America at the Mexico trench. The northernmost segment of this trench, between the Orozco and Rivera fracture zones, has ruptured in a sequence of five large earthquakes from 1973 to 1985; the Jan. 30, 1973 Colima event (M _s 7.5) at the northern end of the segment near Rivera fracture zone; the Mar. 14, 1979 Petatlan event (M _s 7.6) at the southern end of the segment on the Orozco fracture zone; the Oct. 25, 1981 Playa Azul event (M _s 7.3) in the middle of the Michoacan gap; the Sept. 19, 1985 Michoacan mainshock (M _s 8.1); and the Sept. 21, 1985 Michoacan aftershock (M _s 7.6) that reruptured part of the Petatlan zone. Body wave inversion for the rupture process of these earthquakes finds the best: earthquake depth; focal mechanism; overall source time function; and seismic moment, for each earthquake. In addition, we have determined spatial concentrations of seismic moment release for the Colima earthquake, and the Michoacan mainshock and aftershock. These spatial concentrations of slip are interpreted as asperities; and the resultant asperity distribution for Mexico is compared to other subduction zones. The body wave inversion technique also determines theMoment Tensor Rate Functions; but there is no evidence for statistically significant changes in the moment tensor during rupture for any of the five earthquakes. An appendix describes theMoment Tensor Rate Functions methodology in detail.The systematic bias between global and regional determinations of epicentral locations in Mexico must be resolved to enable plotting of asperities with aftershocks and geographic features. We have spatially shifted all of our results to regional determinations of epicenters. The best point source depths for the five earthquakes are all above 30 km, consistent with the idea that the down-dip edge of the seismogenic plate interface in Mexico is shallow compared to other subduction zones. Consideration of uncertainties in the focal mechanisms allows us to state that all five earthquakes occurred on fault planes with the same strike (N65°W to N70°W) and dip (15±3°), except for the smaller Playa Azul event at the down-dip edge which has a steeper dip angle of 20 to 25°. However, the Petatlan earthquake does prefer a fault plane that is rotated to a more east-west orientation—one explanation may be that this earthquake is located near the crest of the subducting Orozco fracture zone. The slip vectors of all five earthquakes are similar and generally consistent with the NUVEL-predicted Cocos-North America convergence direction of N33°E for this segment. The most important deviation is the more northerly slip direction for the Petatlan earthquake. Also, the slip vectors from the Harvard CMT solutions for large and small events in this segment prefer an overall convergence direction of about N20°E to N25°E.All five earthquakes share a common feature in the rupture process: each earthquake has a small initial precursory arrival followed by a large pulse of moment release with a distinct onset. The delay time varies from 4 s for the Playa Azul event to 8 s for the Colima event. While there is some evidence of spatial concentration of moment release for each event, our overall asperity distribution for the northern Mexico segment consists of one clear asperity, in the epicentral region of the 1973 Colima earthquake, and then a scattering of diffuse and overlapping regions of high moment release for the remainder of the segment. This character is directly displayed in the overlapping of rupture zones between the 1979 Petatlan event and the 1985 Michoacan aftershock. This character of the asperity distribution is in contrast to the widely spaced distinct asperities in the northern Japan-Kuriles Islands subduction zone, but is somewhat similar to the asperity distributions found in the central Peru and Santa Cruz Islands subduction zones. Subduction of the Orozco fracture zone may strongly affect the seismogenic character as the overlapping rupture zones are located on the crest of the subducted fracture zone. There is also a distinct change in the physiography of the upper plate that coincides with the subducting fracture zone, and the Guerrero seismic gap to the south of the Petatlan earthquake is in the wake of the Orozco fracture zone. At the northern end, the Rivera fracture zone in the subducting plate and the Colima graben in the upper plate coincide with the northernmost extent of the Colima rupture zone.